Introduction
Welcome to my repository! The topic of this corpus that is game music with a focus on theme music. Over the years games have been evolving not only in the fields of graphics or game mechanics but also in the field of music. In the early stages of videogames, in the eighties, games could not process more than three notes meaning it was not possible to write a tune with complex multi-layered melodies. Over time this changed. It is interesting to look at how this progressed and what the significant changes are. As you can probably tell there is a huge amount of categories out there. It is possible to make categories based on time period or game genre or you can make distinctions based on the music itself. What is also interesting to note is that the popular video games in the beginning were mainly Japanese such as Mario and Final Fantasy. So it could also be a fun project to compare Western and Japanese music. I think that in the beginning because of the classic three notes it is easier to identify game music but later on there would be a blurry distinction between regular music and game music. It is interesting to note that there is a grouping towards a very low valence and a very high intrumentalness. This means that the music in this corpus, in general, is pretty negative and uses almost no voices.
Visualisation
It is interesting to note that there is a grouping towards a very low valence and a very high intrumentalness. This means that the music in this corpus, in general, is pretty negative and uses almost no voices.
Timbre
self-similarity matrix
Comparing Mario theme songs
*** Here we compared the classic Super Mario Bros theme song with the
Super Marion 64 one. The song is the same. The only difference is that
in the first one is in 8-bit. As you can see this makes a big difference
because according to this audio-analysis these two are not similar at
all.